@article{adaptivevococomplexityawarevisualtokenco,
  title = {Adaptive-VoCo: Complexity-Aware Visual Token Compression for Vision-Language Models},
  author = {Xiaoyang Guo and Keze Wang},
  year = {2025},
  eprint = {2512.18496},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2512.18496},
}